Attenuation of STAT3 Phosphorylation Promotes Apoptosis and

Chemosensitivity in Human Osteosarcoma Induced by Raddeanin A

Zhuoying Wang1, 2, 3, Chongren Wang1, 2, 3, Dongqing Zuo1, 2, 3, Tao Zhang2, Fei Yin1, 2,

Zifei Zhou2, Hongsheng Wang2, Jing Xu1, 2, Min Mao2, Gangyang Wang2, Wei Sun1, 2*,

Yingqi Hua1, 2* Zhengdong Cai1, 2*

1 Department of Orthopaedics, Shanghai General Hospital, School of Medicine

Shanghai Jiao Tong University, Shanghai, China;

2 Shanghai Bone Tumor Institution, Shanghai, China;

3 These authors contributed equally to this work.

*Correspondence to: Yingqi Hua, E-mail: yhua@shsmu.edu.cn；Or Wei Sun, Email: viv-sun@163.com

Or Zhengdong Cai, Email: caizhengdong@sjtu.edu.cn

AbstractOsteosarcoma (OS) is the most common primary bone malignancy in adolescents.

One major obstacle for current OS treatment is drug-resistance. Raddeanin A (RA), an

oleanane-type triterpenoid saponin, exerts anti-tumor effects in several tumor models,

but the effect of RA in human drug-resistant OS remained to be elucidated. In the

present study, we investigated the anti-tumor effects of RA in both drug-sensitive and

drug-resistant OS cells and its underlying mechanism. RA inhibited cell proliferation

and colony formation and induced apoptotic cell death in a dose-dependent manner in

both drug-sensitive and drug-resistant cells. Moreover, RA exposure resulted in the

inhibition of interleukin-6 (IL-6)-induced JAK2/STAT3 signaling pathway activation

and target gene expression in both drug-sensitive and drug-resistant cells. Meanwhile,

we observed significantly increased MDR1 and STAT3 expression in drug-resistant

OS cells compared with parental cells. STAT3 overexpression promoted chemo-

resistance and MDR1 protein expression in both drug-sensitive OS cells and drug-

resistant OS cells, while inhibiting STAT3 with siRNA sensitized OS cells to

doxorubicin treatment. In addition, RA synergistically increased doxorubicin toxicity

by increasing its cellular uptake, ablating efflux and downregulating MDR1 in drug-

resistant cells with attenuation of STAT3 Phosphorylation. Finally, RA suppressed in

vivo tumor growth and induced apoptosis in nude mouse using drug-resistant OS tibia

orthotopic model. Taken together, RA is a promising potential therapeutic for the

treatment of doxorubicin resistance in OS.

KEYWORDS: MDR1; Raddeanin A; osteosarcoma; STAT3; drug-resistance

Introduction

Osteosarcoma (OS) is the most common primary bone sarcoma, affecting 1-3 people

per million, and the estimated 5-year overall survival rate for OS patients is

approximately 60%-70% [1]. One major obstacle for current OS treatment is drug

resistance, either intrinsic or acquired, to OS chemotherapeutic agents such as

doxorubicin, ifosfamide and cisplatin, leading to the recurrence of malignant tumors

and, ultimately, patient relapse or death. Drug resistance is widely considered a poor

prognostic indicator for patients with OS [2, 3]. To improve the efficacy of current

chemotherapy, strategies to reverse OS drug resistance have been studied extensively

over the past few decades. Members of the ATP-binding cassette transporter

superfamily, including P-glycoprotein (MDR1) and multidrug resistance-associated

protein 1 and 2 (MRP1/ABCC1 and MRP2/ABCC2), mediate the classical drug

resistance mechanism and play crucial roles in OS resistance to doxorubicin. Among

these proteins, MDR1 was shown to contribute to the tumor response to

chemotherapy in an Asian population, especially in OS and breast cancer, and to

patient prognosis in our previous study [4].

Signal transducer and activator of transcription-3 (STAT3) is primarily activated

and regulated by interleukin-6 (IL-6) family cytokine receptor-associated Janus

kinases (JAKs), and in turn, STAT3 regulates the transcriptional activation of several

anti-apoptotic and pro-proliferative gene products, such as cyclins, B-cell lymphoma-

2 (Bcl-2) and surviving [5]. We have previously shown that inhibition of the STAT3

pathway induces apoptotic cell death and blocks tumor growth in vitro and in vivo in

OS [6-8]. Constitutive activation of STAT3 has been shown to confer resistance to

chemotherapy-induced apoptosis in some malignancies [9-11]. Tang et al [12]

confirmed that STAT3 activation by IL-6 regulates mesenchymal stem cells (MSC)-

induced chemo-resistance and reported that blockade of STAT3 signaling re-sensitized

drug-resistant OS Saos-2 cells to drug treatment. Duan et al [13] found that inhibiting

the STAT3 pathway induces drug-resistant OS cell apoptosis. Thus, STAT3 may be a

promising therapeutic target for overcoming drug resistance in OS. Some researchers

[14, 15] have shown that STAT3 could participate in regulating the transcription of

MDR1 and MDR1 could be a downstream target of STAT3. But the underlying

mechanism is still need to be elucidated.

In our previous study, we have identified that ursolic acid (UA) derivative as

potent anti-tumor agent for OS in preclinical studies [16, 17]. In this study, we show

that Raddeanin A (RA), which shares similar active constituents with UA, also with

anti-tumor activity in several tumor models [18-23], as a JAK/STAT3 pathway

inhibitor in OS. Here we show RA could inhibit tumor proliferation and growth and

induce apoptosis by modulating the STAT3 pathway and downstream target gene

expression in both doxorubicin-sensitive and doxorubicin-resistant OS. Furthermore,

RA synergistically increases doxorubicin toxicity in drug-resistant OS cells by

inhibiting the STAT3/MDR1 signaling axis in vitro and in vivo.

Materials and Methods

Cell lines and culture

The human OS MG63, U2OS, HOS cell lines were obtained from American Type

Culture Collection (ATCC). The human OS drug-resistant U2OSR and KHOSR cell

lines were kind gift from Dr. Duan zhenfeng [24] (Sarcoma Biology Laboratory,

Massachusetts General Hospital, Harvard University). All cell lines were maintained

in DMEM supplemented with 10% FBS and 1% penicillin/streptomycin in a

humidified incubator at 37°C and 5% CO2.

Drugs and antibodies

RA (99.9% purity) was purchased from Yuanye Biotechnology, China, and dissolved

in dimethyl sulfoxide (DMSO) to a 10mM stock solution that was stored in aliquots in

the dark at -20°C. The following antibodies were used for immunoblotting: rabbit

anti-actin (Santa Cruz Biotechnology, CA, USA) and anti-PCNA, anti-caspase-3, anti-

Bcl-xl, anti-Bcl-2, anti-PARP, anti-STAT3, anti-phospho-STAT3 (Tyr705), anti-JAK2,

anti-phospho-JAK2 (Tyr1007/1008), anti-Src, anti-phospho-Src (Tyr416), anti-

MDR1, anti-MRP1 (Cell Signaling Technology, Inc., Danvers, MA, USA). Human

IL-6 was purchased from Sigma (Sigma-Aldrich, Inc., MO, USA).

CCK8 cell viability assay

Cells were seeded into 96-well plates and treated with RA at the indicated

concentrations for 48 h. Cells incubated with 0.1% DMSO in DMEM-h served as the

vehicle control group. CCK8 (20 μl; 5 mg/ml, Dojindo Molecular Technologies, Inc.)

was added to each well, and the plates were incubated for another 4 h at 37 °C

according to the manufacturer’s protocol. Then, the absorbance at 490 nm was

measured using an ELX800 Micro Plate Reader (Bio-Tec Instruments, Inc.). Excel

FORECAST (predict x based on known x and y values) was used to predict the drug

IC50[25]. Three independent experiments were performed in triplicate.

Cell clonogenic assay

Colony formation assays were conducted as previously described [6]. Cells were

plated at 500-1000 cells/well in six-well plates. After vehicle or RA treatment for 48

h, cells were maintained in fresh medium for 10-14 days until visible colonies were

observed. Colonies were then washed, fixed, and stained with 0.1% crystal violet. The

colony number was counted manually. Images were acquired with a digital camera.

Flow cytometry analysis

Cells were cultured in six-well plates (2.5×105/well) and treated as indicated for 48 h.

Apoptosis was detected and then analyzed using FlowJo software as previously

described [26]. Dual parameter dot plots of Annexin V-FITC and PI staining revealed

live cells in the lower-left quadrant (Annexin V-/PI-), early apoptotic cells in the

lower-right quadrant (Annexin V+/PI-), late apoptotic cells in the upper-right quadrant

(Annexin V-/PI+), and necrotic cells in the upper-left quadrant (Annexin V+/PI+).

Immunoblotting

Cells were washed twice with cold 1× PBS solution and lysed with RIPA lysis buffer

(Beyotime, Shanghai, China) containing phosphatase and protease inhibitors

(Beyotime, Shanghai, China). Equivalent amounts of total protein (30-60 μg) were

separated in 8% or 12% polyacrylamide gels and transferred to nitrocellulose filter

membranes. Membranes were blocked for 60 min with 5% milk in PBST (PBS and

0.1% Tween 20), incubated overnight at 4°C with 1:1000 dilutions of the primary

antibodies, washed three times for 10 min each with PBST, and incubated for 1 h with

the appropriate peroxidase-conjugated secondary antibody (1:5000 dilution).

Membranes were then washed with PBST three times for 10 min each and then

developed using the Odyssey two color infrared laser imaging system. The signal

generated by β-actin was used as an internal control.

Reverse transcription polymerase chain reaction (RT-PCR)

RNA was collected and purified from treated cells using Trizol reagent (Invitrogen).

RNA samples (1 μg) were subjected to RT-PCR using the TaKaRa RT-PCR kit

(Takara, Shiga, Japan). Each 10 µL reaction contained 25 ng of sample cDNA. The

following primer sequences were used: GAPDH F 5′-ATG TTC GTC ATG GGT GTG

AA-3′, R 5′-TGT GGT CAT GAG TCC TTC CA-3′; and MDR1 F1, 5′-GGA AGA

CAT GAC CAG GTA TGC C-3′, MDR1 R1, 5′-GGA GAC ATC ATC TGT AAG TCG

GG-3′.

Drug uptake and efflux assay

Doxorubicin cellular uptake was evaluated based on the cellular distribution of red

auto-fluorescence of the drug. Briefly, drug-resistant U2OSR cells (1×105 cells per

well) were cultured in 12-well plates for 24 h. Triplicate wells were treated with

indicated doxorubicin and the indicated concentration of RA for 6 h, and images were

then acquired with a fluorescence microscope (Leica). Drug efflux assays were

performed as described [24]. Drug-sensitive and drug-resistant cells (1×105 cells per

well) were cultured in 12-well plates for 24 h. Triplicate wells were treated with the

indicated concentration of RA for 6 h and then exposed to 50 nM calcein AM. After a

30-min incubation, the cells were washed and fixed with 4% paraformaldehyde. The

nuclei were stained with 4’,6-diamidino-2-phenylindole (DAPI). Images were

acquired with a fluorescence microscope (Leica). Cells were rinsed with DPBS, and

total fluorescent emission in each well was measured with a SpectraMax® M5/M5e

plate reader (Molecular Devices, US).

STAT3 transfection

OS cells were plated at a density of 1.5×105 cells per well in 6-well plates and allowed

to attach overnight. Then, the cells were transfected with 4 μg of pcDNA3 or

pGMSTAT3 using Lipofectamine 2000 (Invitrogen, MA, USA) for 48 h according to

the manufacturer’s protocol, transfected cells were then treated with 2 µg/ml

puromycin for another 24-48 h and used in the following experiments.

Nude mouse tibia orthotopic tumor model

Four-week-old female BALB/c nude mice were housed under standard conditions

with a 12-h light-dark cycle and were fed sufficient water and food. All the animal

procedures were performed in accordance with a protocol approved by the Animal

Care and Use Committee of Shanghai General Hospital and Shanghai Jiaotong

University. KHOSR cells (106) were injected into the medullary cavity of the right

tibia to establish an orthotopic OS model. Two weeks after tibial injection, each

mouse in the RA groups received a weight-based dose of drug by intraperitoneal

injection (ip) every 3 days. For the drug resistance orthotopic mouse models, 20 mice

were randomized to four groups: vehicle (DMSO) (n=5), 5 mg/kg RA (n=5), 1 mg/kg

doxorubicin (n=5), or 5 mg/kg RA plus 1 mg/kg doxorubicin (n=5). Mice in the

vehicle group were injected with 100 µl of PBS containing 10% DMSO on the same

schedule. After 7-10 treatments, the mice were euthanized. Tumors were dissected and

stored in liquid nitrogen or fixed in formalin for further analysis.

TUNEL assay

Apoptosis in tumor samples was identified using a TUNEL Assay Kit (Beyotime,

Beijing, China) according to the manufacturer’s instructions. In brief, paraffin-

embedded slides were deparaffinized with xylene, rehydrated with ethanol and

incubated with proteinase K for antigen retrieval. After several washes with PBS, the

sections were incubated with freshly prepared TUNEL reaction mixture for 1 h at 37

°C in a humidified chamber. Apoptotic cells in randomly chosen fields on the slides

were observed using a light microscope (Leica).

Statistical analysis

Data are presented as the mean ± S.D. Student’s t-test was used to compare two

groups (P<0.05 was considered significant) unless otherwise indicated. All

experiments were performed at least three times.

Results

RA suppresses cell proliferation and colony formation in chemosensitive and

chemoresistant OS cell lines

To evaluate the anti-proliferative effect of RA, we performed CCK8 viability assays

using a panel of OS cells. CCK8 cell viability assays conducted 48 h after vehicle

(0.1% DMSO) or RA treatment showed a dose-dependent inhibition of cell survival in

OS cell lines (Fig. 1A, B). As shown in Fig. 1B, the 48 h IC50 values for HOS,

U2OS, MG63, cells were 1.512 ± 0.034, 3.527 ± 0.018, 3.977 ± 0.055, 3.584 ± 0.045

µM, respectively. The corresponding IC50 values for KHOSR and U2OSR cells were

2.053 ± 0.086 and 6.510 ± 0.062 µM, respectively. The IC50 values of RA were

generally higher in the second group chemoresistant cells than in the first group

corresponding sensitive cells. Colony formation assays are commonly used to study

the survival and proliferation of adherent cells. In the current study, significantly

fewer colonies were observed after HOS and KHOSR OS cells were treated with the

indicated of RA, confirming the inhibitory role of RA on OS proliferation (Fig. 1C).

In addition, PCNA activity, a cellular marker of proliferation, was significantly

suppressed after RA treatment for 48 h (Fig. 1D). Cell viability was inhibited in time-

dependent manner after RA treatment using the indicate concentration

(Supplementary Figure 1A), and the expression of PCNA was significantly gradually

decreased after RA treatment in a certain concentration for a serial time

(Supplementary Figure 1B). These results showed that RA inhibits the proliferation of

chemo sensitive and chemo resistant OS cells in a dose- and time-dependent manner.

RA induces apoptosis in chemosensitive and chemoresistant OS cell lines

Apoptosis is a programmed cell death mechanism characterized by depolymerization

of the cytoskeleton, cell shrinkage, chromatin condensation, nuclear fragmentation

and translocation of phosphatidylserine to the cell surface. It is a common mechanism

by which anti-cancer drugs kill cancer cells. As shown in Fig. 2A, RA induced

significant apoptosis in both HOS, U2OS, and HOSR, U2OSR cells using Flow

cytometry analysis. Moreover, RA markedly activated PARP cleavage and decreased

the expression of the anti-apoptotic genes Bcl-2 and Bcl-xl in a dose dependent

manner (Fig. 2B). And the proapoptotic gene Bax, a member of the Bcl-2 gene

family, was decreased after RA treatment in a dose dependent manner (Fig. 2B).

These data imply that RA induces dose -dependent apoptosis in chemosensitive and

chemoresistant OS cell lines.

RA inhibits interleukin-6 (IL-6)-induced STAT3 Tyr705 phosphorylation in

chemosensitive and chemoresistant OS cell lines

Cytokines such as IL-6, IL-21, IL-23 and IL-27 activate the STAT3 pathway, and the

stimulation of IL-6R leads to the activation of several transcription factors, most

notably STAT3[27]. Our results showed that RA treatment blocked not only p-STAT3

Tyr705 levels but also p-JAK2 Tyr1007/1008 expression in HOS and KHOSR OS cells

(Fig. 3A, B) in dose- and time- dependent manner. To further determine the upstream

pathways involved, we treated OS cells with the indicated concentration of RA with

or without 20 µg/ml IL-6. As shown in Fig. 3C, IL-6 increased the both p-JAK2

Tyr1007/1008 and p-STAT3 Tyr705 levels, which could be quenched by subsequent RA

treatment (Fig. 3C). Moreover, we established a STAT3-overexpressing OS cell line,

HOS-STAT3 and KHOSR-STAT3. As shown in Fig. 3D, overexpression of STAT3

increased p-STAT3 Tyr705 levels by augmenting total STAT3 levels, which could also

be inhibited by RA treatment. These data indicate that RA inhibits STAT3 Tyr705

phosphorylation and target gene expression in both chemosensitive and

chemoresistant OS cells.

RA suppresses MDR1 expression through inhibiting STAT3 in chemoresistant

OS cells

Recent studies have shown that inhibiting STAT3 effectively enhances multidrug

sensitivity by blocking STAT3-mediated MDR1 gene expression in both leukemia and

breast cancer cells [14,15], indicating that MDR1 could be a downstream target of

STAT3. we examined doxorubicin resistance related ABC family genes and STAT3

expression in U2OSR, KHOSR compared with its parental cells. Our results showed

markedly elevated expression of MDR1, MRP1, STAT3Tyr705 phosphorylation and

STAT3 expression in chemoresistant cells compared with the corresponding parental

OS cells (Fig.4A). Then to investigate the possible mechanism, we explored how

MDR1 expression is affected by RA treatment. After being incubated with the

indicated RA for 48 h, KHOSR and U2OSR cells exhibited a dose-dependent

decrease in MDR1 and MRP1 expression (Fig. 4B). To further investigated the

MDR1 expression in level of mRNA, we used real-time PCR assay, and then found

that MDR1 mRNA expression was decreased with RA treated in two human chemo-

resistant OS cells (Fig. 4C).

Surprisingly, overexpression of STAT3 increased MDR1 expression in chemosensitive

MG63 and chemoresistant U2OSR OS cells (Fig. 4D), and the increased STAT3Tyr705

phosphorylation and MDR1 expression mediated by STAT3 overexpression could be

abated by RA treatment again (Fig. 4E). Moreover, STAT3 siRNA and RA treatment

synergistically downregulated MDR1 expression in U2OSR cells (Fig. 4F).

Further CCK8 proliferation assays showed overexpression of STAT3 increased

doxorubicin resistance in MG63 cells, whereas STAT3 siRNA sensitized MG63 cells

to doxorubicin treatment (Fig. 4G). Taken together, those results imply that RA could

suppressed MDR1expression in level of protein and mRNA through inhibiting the

STAT3 activation in chemicoresistance OS cells to restore the drug sensitivity.

RA restores sensitivity to doxorubicin in chemo-resistant OS cells

Drug resistance is a major obstacle to the successful treatment of OS with first-line

chemotherapy such as doxorubicin, cisplatin and methotrexate. Herein, two

chemoresistant OS cell lines were used to explore the effects and mechanism of RA

on OS drug resistance. As shown in Fig. 3A and B, RA exposure induced dose- and

time- dependent apoptosis in chemoresistant cell lines, as well as we observed in

chemosensitive OS cells (Fig. 2). Moreover, U2OSR cells exhibited less doxorubicin

uptake after been treated with 5 µM doxorubicin for 2 h, evaluated by fluorescence

microscopy imaging, which showed cellular red fluorescence in drug-resistant cells

compared with its parental U2OS cell line (supplementary Figure 2). We nextly

measured the efflux of the non-toxic live cell dye calcein AM and the uptake of

doxorubicin in drug-resistant cells upon RA treatment for 2 hours. Calcein AM was a

known substrate of the multidrug resistance protein P-glycoprotein (MDR1/ABCB1)

or other ABC membrane pump proteins, and it has been employed for fluorescent

substrate efflux assays. RA significantly increased the accumulation of fluorescent

calcein, which hydrolyzed from the retainment of intracellular calcein AM by

intracellular esterase, indicating RA could decrease drug efflux in drug-resistant cells

U2OSR (Fig. 5A). Meanwhile RA promoted doxorubicin uptake by facilitating the

red fluorescence doxorubicin transportation in U2OSR cells (Fig. 5A). As shown in

supplementary Figure 3, the accumulation of fluorescent calcein (green) demonstrated

the decrease of calcein AM efflux, and the accumulation of fluorescent doxorubicin

(red) demonstrated the increase of doxorubicin uptake in U2OSR treated with RA in a

dose-dependent manner.

To further verify this finding, we examined the combination effect of doxorubicin in a

serial concentration of 0, 0.1, 0.2, 0.5, 1, and 2 µM (or 0, 0.2, 0.5, 1, 2 and 5 µM) with

RA in drug resistance OS cell KHOSR (or in U2OSR) using CCK8 viability assays

(Fig. 5B). The serial concentration of doxorubicin was 0, 0.05, 0.1, 0.2, 0.5, 1, and 2

µM in KHOSR (and the serial concentration was 0, 0.1, 0.2, 0.5, 1, 2, and 5 µM in

U2OSR), plusing RA with indicated concentration, to analyze the respective IC50 of

doxorubicin plusing with RA of the indicated concentration using the CalcuSyn

system. As shown in Fig.5C, the IC50s of doxorubicin alone in both KHOSR and

U2OSR cell lines were 1.820 ± 0.071 and 3.202 ± 0.083 µM, whereas the IC50 plus

RA (0.2 µM) was 0.630 ± 0.141 µM in KHOSR and in U2OSR the IC50 plus RA (0.5

µM) was 1.256 ± 0.092 µM. RA was capable of reversing drug resistant at

concentrations 19.36- and 57.17-fold lower than that required for doxorubicin in

KHOSR and U2OSR respectively.

To demonstrate the combination effect, we investigated the apoptotic cells using Flow

cytometry analysis and immunoblotting assay. The percent of apoptotic cells was

significantly higher in 0.5 µM doxorubicin combined with 2 µM RA than control or in

either 0.5 µM doxorubicin or 2 µM RA alone (Fig. 5D). As showed in Fig. 5E, the

expression of MDR1 and MRP1 was significantly reduced by RA, and when

combined with RA, the expression of cleaved PARP was increased compared with

doxorubicin or RA respectively. In short, those results indicate RA could modulate the

drug uptake and efflux, induces apoptosis and sensitizes chemo-resistant cells to

doxorubicin treatment in human OS cells.

RA inhibits tumor growth in orthotopic chemoresistant OS animal model

To assess whether the biologic effect of RA on chemo-resistant OS is potentially

effective in vivo, we established a nude mouse orthotopic model using chemo-

resistant KHOSR cells as described previously [29]. Mice were randomized into four

groups and then treated by ip injection with vehicle, 5 mg/kg RA, 1 mg/kg

doxorubicin and RA plus doxorubicin. As shown in Fig. 6A, 5 mg/kg RA, 1 mg/kg

doxorubicin or RA plus doxorubicin significantly decreased tumor weight compared

with vehicle. Interestingly, RA showed a significant synergistic effect with

doxorubicin, which correlated with the in vitro findings as we indicated in Fig5B, and

5D. However, there were no differences in mouse body weight, indicating that RA

treatment have tolerable toxicity in vivo (Fig. 6C). The apoptosis index in tumor

samples was analyzed by TUNEL assay. In agreement with the in vitro study finding,

treatment with RA plus doxorubicin caused significantly more apoptosis than the

other treatments (Fig. 6B). Furthermore, RA downregulated STAT3Tyr705

phosphorylation and MDR1 expression in tumor samples (Fig. 6D). These results

indicate that RA inhibits in vivo tumor growth in an orthotopic chemoresistance model

of human OS.

Discussion

The introduction of biologic agents and the use of additional cytotoxic chemotherapy

has not definitively improved the survival of patients with OS in the past two decades.

Therefore, there is an unmet need to discover more effective agents to overcome

chemotherapy drug resistance. Our previous studies found that oleanolic acid

derivates or analogues exert anti-tumor effect in OS preclinical models [16, 17], and

might be potential new drug for OS clinical trials. While our other studies support the

hypothesis that activated STAT3 might be target for anti-tumor therapy and natural

compounds targeting the STAT3 pathway could be used for inhibiting OS [6-8].

Herein, we identified RA, a triterpenoid saponin extracted from the root of Anemone

raddeana Regel, which shares similar active form with Oleanolic acid, as effective

anti-tumor agent in OS, and the anti-proliferative and pro-apoptotic effects of RA

could be modulated by the STAT3 pathway through its target genes regulation.

Our study showed that STAT3 expression is higher in chemoresistant cell lines

(KHOSR and U2OSR) than its parental cells. Moreover, RA, attenuated MDR1

mRNA and protein expression in a dose-dependent manner in all the tested OS cell

lines. Thus, it is possible that STAT3 contributes to doxorubicin resistance in OS.

Suppressing STAT3 increased the sensitivity of chemo-resistant cells to doxorubicin

by increasing its uptake and ablating drug efflux in human OS. These finding

prompted the discovery that inhibiting STAT3 could effectively inhibit tumor

progression and reverse doxorubicin resistance in OS. Yun et al [30] found that

cinnamaldehyde derivative (CB-PIC) sensitizes drug-resistant cancer cells to drug-

induced apoptosis by suppressing MDR1, although CB-PIC did not directly attenuate

MDR1 activity but rather inhibited MDR1 mRNA and protein expression by

suppressing the STAT3 and AKT signaling pathways. While the study by Zhang et al

[14] showed that STAT3 could bind the +64~+72 region of the MDR1 promoter and

thus initiate its transcription in leukemia cells, the blockade of STAT3 activation by a

STAT3 decoy oligodeoxynucleotide (ODN) promoted adriamycin accumulation and

increased the cellular sensitivity to adriamycin by downregulating the transcription

and protein expression of MDR1. In our study, RA inhibited MDR1 mRNA and

protein expression in a dose-dependent manner, and the overexpression of STAT3

promoted doxorubicin resistance and MDR1 expression in chemo-sensitive MG63

cells and chemo-resistant U2OSR cells. Moreover, STAT3 siRNA and RA treatment

synergistically downregulated MDR1 expression in U2OSR cells. Hence, we

concluded that MDR1 is a downstream target of STAT3 in OS. A recent study by Peng

et al [31] showed RA could reverse STAT3/NFIL3 mediated chemotherapy resistance

by methotrexate (MTX), 5-FU and etoposide (VP16) through inducing apoptosis in

choriocarcinoma cells. They have identified STAT3/NFIL3 axis as a new mechanism

for choriocarcinoma drug resistance, but the weakness of their study is they only

presented the in vitro proof without showing the reversal effect in animal models.

Whereas our study providing strong evidence for the interaction of STAT3 and

MDR1, indicating inhibition of STAT3/MDR1 axis by RA might be key to reverse

doxorubicin induced chemo-resistance in OS.

Regarding the interplay between STAT3, its target genes and MDR1, Ji et al [32]

showed that a novel triazolonaphthalimide derivative LSS-11 could hinder the binding

of STAT3 to the MDR1 and MRP1 promoters by conducting chromatin

coimmunoprecipitation (ChIP) assay. Indicating MDR1 is regulated by the STAT3 at

the transcription level. Our study provide evidence that STAT3 regulate MDR1

expression in vitro and in vivo. However, how STAT3 target genes are associated with

MDR1 function remains to be determined. Thus, more efforts are still needed to

elucidate the possible mechanisms in the future.

Abbreviations

OS: osteosarcoma; RA: Raddeanin A; STAT3: Signal transducers and activators of

transcription; MDR1: multi-drug resistance 1; MRP1: Multidrug Resistance-

Associated Protein;

Conflicts of interest

The author(s) indicated no potential conflicts of interest including employment,

consultancies, stock ownership, honoraria, paid expert testimony, patent

applications/registrations, and grants or other funding.

Acknowledgements.

We thank animal Center of Shanghai General Hospital, for all the animal holding and

care. This project was supported by the NSFC (81502604, 81501584, and 81702973),

the Shanghai Science and Technology Commission (14140904000), the Doctoral

Innovation Fund of Shanghai Jiaotong University School of Medicine (No.

BXJ201732), the Shanghai Municipal Commission of Health and Family Planning

(No. 20164Y0270), and a Research Grant from the Shanghai Hospital Development

Center (SHDC12013107).

References

1. Isakoff MS, Bielack SS, Meltzer P, Gorlick R. Osteosarcoma: Current Treatment and a Collaborative Pathway to Success. J Clin Oncol. 2015; 33(27):3029-3035.

2. Li S, Sun W, Wang H, Zuo D, Hua Y, Cai Z. Research progress on the multidrug resistance mechanisms of osteosarcoma chemotherapy and reversal. Tumour Biol. 2015; 36(3):1329-1338.

3. Serra M, Pasello M, Manara MC, Scotlandi K, Ferrari S, Bertoni F, Mercuri M, Alvegard TA, Picci P, Bacci G et al. May P-glycoprotein status be used to stratify high-grade osteosarcoma

patients? Results from the Italian/Scandinavian Sarcoma Group 1 treatment protocol. Int J Oncol. 2006; 29(6):1459-1468.

4. Zhou Z, Chen Q, Zuo D, Wang H, Hua Y, Cai Z. ABCB1 (rs1128503) polymorphism and response to chemotherapy in patients with malignant tumors-evidences from a meta-analysis. Int J Clin Exp Med. 2015; 8(1):265-272.

5. Zhao C, Li H, Lin HJ, Yang S, Lin J, Liang G. Feedback Activation of STAT3 as a Cancer Drug-Resistance Mechanism. Trends Pharmacol Sci. 2016; 37(1):47-61.

6. Zuo D, Zhou Z, Wang H, Zhang T, Zang J, Yin F, Sun W, Chen J, Duan L, Xu J et al. Alternol, a natural compound, exerts an anti-tumour effect on osteosarcoma by modulating of STAT3 and ROS/MAPK signalling pathways. J Cell Mol Med. 2017; 21(2):208-221.

7. Zhang T, Li J, Yin F, Lin B, Wang Z, Xu J, Wang H, Zuo D, Wang G, Hua Y et al. Toosendanin demonstrates promising antitumor efficacy in osteosarcoma by targeting STAT3. Oncogene. 2017; 36(47):6627-6639..

8. Zhang T, Li S, Li J, Yin F, Hua Y, Wang Z, Lin B, Wang H, Zou D, Zhou Z et al. Natural product pectolinarigenin inhibits osteosarcoma growth and metastasis via SHP-1-mediated STAT3 signaling inhibition. Cell death & disease. 2016; 7(10):e2421.

9. Peng J, Wang Q, Liu H, Ye M, Wu X, Guo L. EPHA3 regulates the multidrug resistance of small cell lung cancer via the PI3K/BMX/STAT3 signaling pathway. Tumour Biol. 2016; 37(9):11959-11971.

10. Zhang F, Wang Z, Fan Y, Xu Q, Ji W, Tian R, Niu R. Elevated STAT3 Signaling-Mediated Upregulation of MMP-2/9 Confers Enhanced Invasion Ability in Multidrug-Resistant Breast Cancer Cells. Int J Mol Sci. 2015; 16(10):24772-24790.

11. Zhu H, Chen X, Chen B, Chen B, Fan J, Song W, Xie Z, Jiang D, Li Q, Zhou M et al. Activating transcription factor 4 mediates a multidrug resistance phenotype of esophageal squamous cell carcinoma cells through transactivation of STAT3 expression. Cancer letters. 2014; 354(1):142-152.

12. Tu B, Zhu J, Liu S, Wang L, Fan Q, Hao Y, Fan C, Tang TT. Mesenchymal stem cells promote osteosarcoma cell survival and drug resistance through activation of STAT3. Oncotarget. 2016; 7(30):48296-48308.

13. Ryu K, Susa M, Choy E, Yang C, Hornicek FJ, Mankin HJ, Duan Z. Oleanane triterpenoid CDDO-Me induces apoptosis in multidrug resistant osteosarcoma cells through inhibition of Stat3 pathway. BMC cancer. 2010; 10:187.

14. Zhang X, Xiao W, Wang L, Tian Z, Zhang J. Deactivation of signal transducer and activator of transcription 3 reverses chemotherapeutics resistance of leukemia cells via down-regulating P-gp. PloS one. 2011; 6(6):e20965.

15. Bourguignon LY, Peyrollier K, Xia W, Gilad E. Hyaluronan-CD44 interaction activates stem cell marker Nanog, Stat-3-mediated MDR1 gene expression, and ankyrin-regulated multidrug efflux in breast and ovarian tumor cells. J Biol Chem. 2008; 283(25):17635-17651.

16. Hua Y, Zhang Z, Li J, Li Q, Hu S, Li J, Sun M, Cai Z. Oleanolic acid derivative Dex-OA has potent anti-tumor and anti-metastatic activity on osteosarcoma cells in vitro and in vivo. Invest New Drugs. 2011; 29(2):258-265.

17. Chen J, Fu H, Wang Z, Yin F, Li J, Hua Y, Cai Z. A new synthetic ursolic acid derivative IUA with anti-tumor efficacy against osteosarcoma cells via inhibition of JNK signaling pathway. Cell Physiol Biochem. 2014; 34(3):724-733.

18. Gu G, Qi H, Jiang T, Ma B, Fang Z, Xu H, Zhang Q. Investigation of the cytotoxicity, apoptosis and pharmacokinetics of Raddeanin A. Oncology letters. 2017; 13(3):1365-1369.

19. Li JN, Yu Y, Zhang YF, Li ZM, Cai GZ, Gong JY. Synergy of Raddeanin A and cisplatin induced therapeutic effect enhancement in human hepatocellular carcinoma. Biochem Biophys Res Commun. 2017; 485(2):335-341.

20. Teng YH, Li JP, Liu SL, Zou X, Fang LH, Zhou JY, Wu J, Xi SY, Chen Y, Zhang YY et al. Autophagy Protects from Raddeanin A-Induced Apoptosis in SGC-7901 Human Gastric Cancer Cells. Evid Based Complement Alternat Med. 2016; 2016:9406758.

21. Guan YY, Liu HJ, Luan X, Xu JR, Lu Q, Liu YR, Gao YG, Zhao M, Chen HZ, Fang C. Raddeanin A, a triterpenoid saponin isolated from Anemone raddeana, suppresses the angiogenesis and growth of human colorectal tumor by inhibiting VEGFR2 signaling. Phytomedicine. 2015; 22(1):103-10.

22. Xue G, Zou X, Zhou JY, Sun W, Wu J, Xu JL, Wang RP. Raddeanin A induces human gastric cancer cells apoptosis and inhibits their invasion in vitro. Biochem Biophys Res Commun. 2013; 439(2):196-202.

23. Wang MK, Ding LS, Wu FE. Antitumor effects of raddeanin A on S180, H22 and U14 cell xenografts in mice. Ai zheng. 2008, 27(9):910-913.

24. Ye S, Zhang J, Shen J, Gao Y, Li Y, Choy E, Cote G, Harmon D, Mankin H, Gray NS et al. NVP-TAE684 reverses multidrug resistance (MDR) in human osteosarcoma by inhibiting P-glycoprotein (PGP1) function. Br J Pharmacol. 2016; 173(3):613-626.

25. Tong ZG, Liu N, Song HS, Li JQ, Jiang J, Zhu JY, Qi JP. Cytochalasin B inhibits the proliferation of human glioma U251 cells through cell cycle arrest and apoptosis. Genet Mol Res. 2014; 13(4):10811-10822.

26. Fu Z, Deng B, Liao Y, Shan L, Yin F, Wang Z, Zeng H, Zuo D, Hua Y, Cai Z. The anti-tumor effect of shikonin on osteosarcoma by inducing RIP1 and RIP3 dependent necroptosis. BMC cancer. 2013; 13:580.

27. Heinrich PC, Behrmann I, Muller-Newen G, Schaper F, Graeve L. Interleukin-6-type cytokine signalling through the gp130/Jak/STAT pathway. Biochem J. 1998; 334 ( Pt 2):297-314.

28. Wang H, Li W, Xu J, Zhang T, Zuo D, Zhou Z, Lin B, Wang G, Wang Z, Sun W et al. NDRG1 inhibition sensitizes osteosarcoma cells to combretastatin A-4 through targeting autophagy. Cell death & disease. 2017; 8(9):e3048.

29. Wang Z, Wang C, Zhou Z, Sun M, Zhou C, Chen J, Yin F, Wang H, Lin B, Zuo D et al. CD151-mediated adhesion is crucial to osteosarcoma pulmonary metastasis. Oncotarget. 2016; 7(37):60623-60638.

30. Yun M, Lee D, Park MN, Kim EO, Sohn EJ, Kwon BM, Kim SH. Cinnamaldehyde derivative (CB-PIC) sensitizes chemo-resistant cancer cells to drug-induced apoptosis via suppression of MDR1 and its upstream STAT3 and AKT signalling. Cell Physiol Biochem. 2015; 35(5):1821-1830.

31. Peng Z, Zhang C, Zhou W, Wu C, Zhang Y. The STAT3/NFIL3 signaling axis-mediated chemotherapy resistance is reversed by Raddeanin A via inducing apoptosis in choriocarcinoma cells. J Cell Physiol. 2018; 233(7):5370-5382.

32. Ji L, Liu X, Zhang S, Tang S, Yang S, Li S, Qi X, Yu S, Lu L, Meng X et al: The Novel Triazolonaphthalimide Derivative LSS-11 Synergizes the Anti-Proliferative Effect of Paclitaxel via STAT3-Dependent MDR1 and MRP1 Downregulation in Chemoresistant Lung

Cancer Cells. Molecules 2017, 22(11): 1822.

Figures

Fig. 1. RA blocks cell proliferation and colony formation in both drug-sensitive

and drug-resistant human OS cells.

(A) The human OS HOS, KHOSR, U2OS, U2OSR and MG63 cell lines were treated

with vehicle (0.1% DMSO) or the indicated concentration of RA for 48 h, and cell

viability was measured by CCK8 assay. (B) IC50 of RA in human OS cell lines at 48

h. (C) OS cells (HOS and KHOSR) were plated in six-well plates and treated with

vehicle or the indicated concentration of RA for 48 h, cells were then maintained in

fresh medium for 10-14 days. Colonies were fixed and stained with 0.1% crystal

violet. (D) HOS, KHOSR, U2OS, and U2OSR cells were treated with the indicated

concentration of RA for 48 h, and the expression of the proliferation marker PCNA

was analyzed by immunoblotting. β-actin was used as a loading control. *P<0.05, **

P<0.01 compared with vehicle control.

Fig. 2. RA induces apoptosis in drug-sensitive and drug-resistant human OS

cells. HOS, KHOSR, U2OS, and U2OSR OS cells were treated with vehicle or the

indicated concentration of RA for 48 h. (A) The apoptosis index was determined by

flow cytometry with Annexin V-FITC and PI staining. (B) expression levels of

cleaved PARP, Bax, Bcl-2 and Bcl-xl were determined by immunoblotting. *P<0.05,

** P<0.01 compared with vehicle control.

Fig. 3. RA inhibits interleukin-6 (IL-6)-induced STAT3 Tyr705 phosphorylation and

target gene expression.

(A, B) drug-sensitive HOS and drug-resistant KHOSR OS cells were treated with the

indicated RA concentration for 48 h or with 2.0 M RA for different periods of time

(0, 12, 24, 36, 48 h). The expression of STAT3, p-STAT3Tyr705, JAK2, p-JAK2Tyr1007/1008,

Src, and p-SrcTyr416 were determined by immunoblotting. (C) HOS and KHOSR OS

cells were treated with the indicated RA concentration for 48 h with or without

subsequent 20 µg/ml IL-6 treatment for 30 min. Then, the levels of STAT3, p-

STAT3Tyr705, JAK2, and p-JAK2Tyr1007/1008 were measured by immunoblotting. (D) HOS

and KHOSR cells were transfected with pCDNA or pGMSTAT3 for 48 h and selected

with 2 g/ml puromycin for 48 h, STAT3 were over expressed in the HOS-STAT3

and KHOSR-STAT3 cells. (E) the resulting HOS-STAT3 compared with HOS and

KHOSR-STAT3 compared with KHOSR cells were then treated with the indicated

concentration of RA, and STAT3 and p-STAT3Tyr705 expression levels were detected by

immunoblotting. β-actin was used as a loading control. *P<0.05, ** P<0.01 compared

with vehicle control.

Fig. 4. RA inhibits the STAT3 phosphorylation and reduces MDR1 expression in

drug-resistant OS.

(A) MDR1, MRP1, STAT3, and p-STAT3Tyr705 levels were determined by

immunoblotting in drug-resistant cells compared with drug-sensitive cells (KHOSR

and HOS, U2OSR and U2OS). (B) KHOSR and U2OSR cells were treated with

vehicle or the indicated concentration of RA for 48 h. MDR1 and MRP1 expression

were detected by immunoblotting. (C) MDR1 expression was determined by real-time

qPCR in KHOSR and U2OSR after RA treated. (D) U2OSR and MG63 cells were

transfected with pCDNA or pGMSTAT3 for 48 h and then treated with puromycin to

establish a STAT3-overexpressing cell line (MG63-STAT3 and U2OSR-STAT3).

MDR1 expression was detected by immunoblotting. (E) STAT3, p-STAT3Tyr705, and

MDR1 levels were determined by immunoblotting in STAT3-overexpressing cell lines

treated with series concentrations of RA. (F) U2OSR cells were treated with the

indicated concentration of RA with or without STAT3 siRNA transfection for 48 h,

and MDR1, STAT3, and p-STAT3Tyr705 expression were determined by

immunoblotting. (G) MG63 cells were transfection with NC/STAT3 siRNA or

pcDNA3/pGMSTAT3 for 48 h, treated with or without doxorubicin for 48h. The cell

viability was detected by CCK8 assay.

Fig. 5 RA reverses doxorubicin resistance in human OS cells by inhibiting STAT3

phosphorylation.

(A) Cells were then treated with the indicated concentration of RA for 2 hours and

then incubated with calcein AM for 30 min, calcein AM efflux was evaluated by green

fluorescence observed using a fluorescence microscope and quantified by

SpectraMax® M5/M5e plate reader. Cells were treated with the indicated

concentrations of RA for 2 hours and doxorubicin, and doxorubicin uptake was

evaluated by red fluorescence observed in fluorescence images and quantified by

SpectraMax® M5/M5e plate reader. The cell nucleuses were stained by DAPI, which

produced blue fluorescence. Relative fluorescence activity meaned the ratio of green

(or red) quantity related to blue quantity. (B) KHOSR and U2OSR cells were treated

with RA in combination with the indicated concentration of doxorubicin for 48 h, and

cell viability was determined by CCK8 assay. (C) IC50 of doxorubicin in presence of

RA (0.5, 1, 2, and 5 µM) were listed in drug- resistant cells KHOSR and U2OSR. (D)

U2OSR cells were treated with or without doxorubicin pretreated with or without of

RA for 2 h and then subjected to Annexin V-FITC/PI staining and flow cytometry

analysis. (E) MDR1, MRP1, STAT3 phosphorylation, total STAT3, and cleaved-PARP

expression were detected by immunoblotting in U2OSR cells treated with doxorubicin

in presence or absence of RA. β-actin was used as a loading control. *P<0.05,

**P<0.01 compared with vehicle control.

Fig. 6. RA inhibits the in vivo growth of drug-resistant OS. (A) Macroscopic

appearance of OS tumors in the tibia of BALB/c nude mice after treatment. Tumor

weight quantification in BALB/c nude mice after treatment. (B) TUNEL assay of

tumor samples from the leg in different treatment groups. (C) Body weight was

measured every 3 days. (D) Tumor samples were lysed and subjected to

immunoblotting with the indicated antibodies. β-actin was used as a loading control.

*P<0.05, **P<0.01 compared with vehicle control.